Move bitstream retrieval process to Item class

* Add separate class methods for getting bitstreams, ids, and metadata
* Add class method to consolidate steps required for creating a DSpace item
* Update test suite

Author: jonavellecuerdo

dsaps/cli.py

@@ -153,14 +153,11 @@ def additems(
             "Option '--metadata-csv' must be used or " "run 'reconcile' before 'additems'"
         )
 
-    bitstream_file_paths = helpers.get_bitstreams_from_csv(metadata_csv)
     dspace_collection = dspace.Collection(uuid=collection_uuid)
 
     with open(metadata_csv, "r") as csvfile:
         metadata = csv.DictReader(csvfile)
-        dspace_collection = dspace_collection.create_metadata_for_items_from_csv(
-            metadata, mapping
-        )
+        dspace_collection = dspace_collection.add_items(metadata, mapping)
 
     for item in dspace_collection.items:
         logger.info(f"Posting item: {item}")
@@ -170,7 +167,7 @@ def additems(
         item.uuid = item_uuid
         item.handle = item_handle
         logger.info(f"Item posted: {item_uuid}")
-        for file_path in bitstream_file_paths.get(item.item_identifier):
+        for file_path in item.bitstreams:
             file_name = file_path.split("/")[-1]
             bitstream = dspace.Bitstream(name=file_name, file_path=file_path)
             logger.info(f"Posting bitstream: {bitstream}")

dsaps/dspace.py

@@ -1,12 +1,16 @@
-import operator
+from __future__ import annotations
 
+import ast
 import attr
+import operator
 import requests
-import smart_open
 import structlog
 
+import smart_open
+
 from attrs import field, define
 
+
 logger = structlog.get_logger()
 op = operator.attrgetter("name")
 
@@ -200,19 +204,39 @@ class Item(Object):
     source_system_identifier = field(default=None)
 
     @classmethod
-    def metadata_from_csv_row(cls, record, mapping):
+    def create(cls, record, mapping) -> Item:
+        return cls(
+            metadata=cls.get_metadata(record, mapping),
+            bitstreams=cls.get_bitstreams(record),
+            **cls.get_ids(record, mapping),
+        )
+
+    @classmethod
+    def get_bitstreams(cls, record) -> list:
+        if bitstreams := record.get("bitstreams"):
+            return ast.literal_eval(bitstreams)
+
+    @classmethod
+    def get_ids(cls, record, mapping) -> dict:
+        ids = {}
+        if item_id_mapping := mapping.get("item_identifier"):
+            ids["item_identifier"] = record.get(item_id_mapping["csv_field_name"])
+        if source_system_id_mapping := mapping.get("source_system_identifier"):
+            ids["source_system_identifier"] = record.get(
+                source_system_id_mapping["csv_field_name"]
+            )
+        return ids
+
+    @classmethod
+    def get_metadata(cls, record, mapping) -> list:
         """Create metadata for an item based on a CSV row and a JSON mapping field map."""
         metadata = []
         for field_name, field_mapping in mapping.items():
+            if field_name in ["item_identifier", "source_system_identifier"]:
+                continue
             field_value = record[field_mapping["csv_field_name"]]
 
             if field_value:
-                if field_name == "item_identifier":
-                    item_identifier = field_value
-                    continue  # file_identifier is not included in DSpace metadata
-                if field_name == "source_system_identifier":
-                    # source_system_identifier = field
-                    continue  # source_system_identifier is not included in DSpace
                 delimiter = field_mapping["delimiter"]
                 language = field_mapping["language"]
                 if delimiter:
@@ -234,22 +258,18 @@ def metadata_from_csv_row(cls, record, mapping):
                             language=language,
                         )
                     )
-        return cls(
-            metadata=metadata,
-            item_identifier=item_identifier,
-            # source_system_identifier=source_system_identifier,
-        )
+        return metadata
 
 
 @define
 class Collection(Object):
     items = field(factory=list)
 
     @classmethod
-    def create_metadata_for_items_from_csv(cls, csv_reader, field_map):
+    def add_items(cls, csv_reader, field_map) -> Collection:
         """Create metadata for the collection's items based on a CSV and a JSON mapping
         field map."""
-        items = [Item.metadata_from_csv_row(row, field_map) for row in csv_reader]
+        items = [Item.create(row, field_map) for row in csv_reader]
         return cls(items=items)
 
 

tests/conftest.py

@@ -35,6 +35,13 @@ def source_metadata_csv():
         yield reader
 
 
+@pytest.fixture
+def source_metadata_csv_with_bitstreams():
+    with open("tests/fixtures/updated-source_metadata.csv") as file:
+        reader = csv.DictReader(file)
+        yield reader
+
+
 @pytest.fixture()
 def dspace_client():
     dspace_client = dspace.DSpaceClient("mock://example.com/")

tests/test_models.py

@@ -78,19 +78,45 @@ def test_build_uuid_list(dspace_client):
     assert "1234" in child_list
 
 
-def test_collection_create_metadata_for_items_from_csv(
-    source_metadata_csv, source_config
-):
-    collection = Collection.create_metadata_for_items_from_csv(
-        source_metadata_csv, source_config["mapping"]
-    )
+def test_collection_add_items(source_metadata_csv, source_config):
+    collection = Collection.add_items(source_metadata_csv, source_config["mapping"])
     assert len(collection.items) == 5
 
 
-def test_item_metadata_from_csv_row(source_metadata_csv, source_config):
+def test_item_create(source_metadata_csv_with_bitstreams, source_config):
+    record = next(source_metadata_csv_with_bitstreams)
+    assert attr.asdict(Item.create(record, source_config["mapping"])) == {
+        "uuid": None,
+        "name": None,
+        "handle": None,
+        "link": None,
+        "type": None,
+        "metadata": [
+            {"key": "dc.title", "value": "Title 1", "language": "en_US"},
+            {"key": "dc.contributor.author", "value": "May Smith", "language": None},
+        ],
+        "bitstreams": ["s3://mocked-bucket/one-to-one/aaaa_001_01.pdf"],
+        "item_identifier": "001",
+        "source_system_identifier": None,
+    }
+
+
+def test_item_get_ids(source_metadata_csv, source_config):
+    record = next(source_metadata_csv)
+    assert Item.get_ids(record, source_config["mapping"]) == {"item_identifier": "001"}
+
+
+def test_item_get_bitstreams(source_metadata_csv_with_bitstreams, source_config):
+    record = next(source_metadata_csv_with_bitstreams)
+    assert Item.get_bitstreams(record) == [
+        "s3://mocked-bucket/one-to-one/aaaa_001_01.pdf"
+    ]
+
+
+def test_item_get_metadata(source_metadata_csv, source_config):
     record = next(source_metadata_csv)
-    item = Item.metadata_from_csv_row(record, source_config["mapping"])
-    assert attr.asdict(item)["metadata"] == [
+    metadata = Item.get_metadata(record, source_config["mapping"])
+    assert [attr.asdict(m) for m in metadata] == [
         {"key": "dc.title", "value": "Title 1", "language": "en_US"},
         {"key": "dc.contributor.author", "value": "May Smith", "language": None},
     ]